DESCRIPTION (Applicant's Abstract): The new and still evolving antiviral agents for HIV have demonstrated early success in reducing viral load and reversing an inexorable clinical course in some patients. The use-effectiveness of these drugs, and their impact on persons and on the AIDS epidemic, will depend not only on pharmacologic and molecular biologic factors (e.g., drug tolerance, longer term side effects, drug resistance) but on economic, cultural, and social factors as well (e.g., access and availability, compliance, changes in risk behavior, changes in the risk configuration of social networks, and the evolution of these factors). Traditionally, clinics are not in a position to assess the latter factors, since they are not structured to make home visits, do outreach, or evaluate patients' social context. In this study, the investigators propose to examine the influence of social network factors on the natural history of antiviral drug use, and on the use-effectiveness of these agents on the health of individuals and on the probability of disease transmission. The role of social network factors (risk configuration, stability, structure, composition, the strength and density) will be considered simultaneously with the role of other critical epidemiological, clinical, and psychological factors. Four groups of persons will be studied: Group I (n=180) -- persons ascertained in the clinic who are placed on antiviral therapy (HIV+); Group II (n=60) -- persons ascertained in the clinic who are potentially eligible for, but are not currently receiving, antiviral therapy (HIV+); Group III (n=60) -- persons ascertained in the community who are infected with HIV, for whom the clinic is available, but who do not use the clinic and are not on antiviral therapy; Group IV (n=60)-- persons ascertained in the community who have demographic and behavioral characteristics similar to those in Group III, but who are not infected with HIV. Information from clinical and laboratory assessments, from psychological interviews and tests, and from field outreach assessment, which includes using both ethnographic and social network analysis techniques to define the social environment, will be collected. Each person will be interviewed at 6-month intervals, for a total of four interviews. The collected information will be used to assess the role of social networks on drug adherence in the presence of potentially confounding epidemiological, clinical, and psychological factors. The study will thus provide information about a variety of factors -- both measured in the clinic and traditionally not available to the clinic -- that can influence the natural history of antiviral drug use, and will serve as a basis for comprehensive assessment of effectiveness and for the design of potential interventions.